to a codebook they have developed. Color coding is another option, as is the
use of computer software, but ultimately researchers come up with their own
systems by which to organize the data into manageable units.
Tesch (1990) believed that traditional hand coding resulted in more data in-
stead of less, and that data reduction is a better alternative. Miles and Huberman
(1994) stated that “data reduction refers to the process of selecting, focusing,
simplifying, abstracting, and transforming the data that appear in written-up
field notes or transcriptions” (p. 10). Data reduction is part of analysis, and the
researcher must make decisions about which data are most representative of
the entire story being told. As researchers choose which patterns or categories
best present the most essential chunks of the data, they become the tool for
data analysis and interpretation.
There is no perfect process for coding data, and there is no wrong process
either. Ganapathy (2016) suggested that coding is essentially a personal filing
system, and this is why there is no exact way to code data. As a result, the
knowledge and skill of the qualitative researcher take on such great significance
and importance in qualitative research.
Traditional open coding (Corbin & Strauss, 1990, 2008) is one way to describe
the first step in grouping the data into categories that seem logical. Researchers
should always leave an audit trail by keeping a log of their ongoing thoughts
and ideas about the naming and assigning of categories to the data so that they
can later examine why certain labels were chosen and others discarded. Axial
coding (Corbin & Strauss, 1990, 2008) takes the analysis process further and
requires researchers to compare the categories and labels, defining and explor-
ing relationships among them.
After the data have been organized by basic coding, Marshall and Rossman
(2011) suggest that the next step is to generate categories, themes, and patterns.
Researchers must search for categories of meaning that best express “relevant
themes, repeating ideas or language, and patterns of belief linking people and
settings.” Codes might be connected through a common theme. There might be
patterns across and within the themes. Researchers look at how codes overlap
and how themes and patterns emerge in relationship to each other. The analysis
process is done with the end result in mind: to stay true to the data and best
represent the story or phenomenon being studied.
After the data are categorized and themes emerge, researchers need to pull
everything together to paint a picture of the underlying truths revealed by the
data. This pulling together of the evidence into a meaningful report that can
be communicated to others is another challenge. As Sandelowski (1995) stated,
“Qualitative analysis also has artistic dimensions that are often inchoate and
incommunicable, involving playfulness, imaginativeness, and creativity” (p. 375).
KEY TERMS
data reduction:
The simplification
of large amounts
of data obtained
from qualitative
interviews or other
sources
open coding:
The grouping of
qualitative data
into categories that
seem logical
axial coding:
The analysis
of categories
and labels after
completion of open
coding14.1 Qualitative Data Analysis 381